diff_of_means ratio_of_sd amplitude_ratio_of_means maximum_error ks_mean_on_coarse_res_with_extremes rainy_hours_ratio_of_means qqplot_mae acf_mae extremogram_mae
nv.mri_esm2_0.ssp370 0.67% 0.899 0.632 0.321 0.565 0.893 0.020 0.158 0.107
lstm.mri_esm2_0.ssp370 2.58% 1.004 0.819 0.361 0.167 1.093 0.013 0.079 0.024
cnn.mri_esm2_0.ssp370 6.34% 0.950 0.782 0.334 0.161 1.136 0.018 0.092 0.037
nv.mri_esm2_0.ssp434 7.34% 0.901 0.612 0.333 0.507 0.950 0.020 0.140 0.094
nv.mri_esm2_0.ssp245 7.74% 0.853 0.609 0.339 0.496 0.924 0.025 0.144 0.086
xgboost.mri_esm2_0.ssp370 8.41% 0.938 0.659 0.278 0.324 1.094 0.015 0.122 0.067
lstm.mri_esm2_0.ssp245 9.04% 0.959 0.775 0.403 0.153 1.133 0.021 0.071 0.020
lstm.mri_esm2_0.ssp434 9.43% 0.977 0.742 0.376 0.223 1.154 0.022 0.086 0.035
nv.cesm2.ssp585 11.56% 0.865 0.603 0.335 0.485 0.968 0.027 0.134 0.099
cnn.mri_esm2_0.ssp434 13.51% 0.942 0.708 0.352 0.202 1.218 0.031 0.090 0.040
cnn.mri_esm2_0.ssp245 13.85% 0.898 0.726 0.353 0.160 1.185 0.030 0.086 0.028
xgboost.ec_earth3.ssp434 -14.57% 0.938 0.781 0.284 0.202 0.805 0.039 0.130 0.045
nv.cesm2.ssp245 15.06% 0.831 0.588 0.329 0.495 0.973 0.032 0.128 0.109
cnn.ec_earth3.ssp434 -15.29% 0.941 0.860 0.264 0.119 0.848 0.039 0.115 0.024
xgboost.mri_esm2_0.ssp245 15.36% 0.886 0.616 0.290 0.293 1.140 0.028 0.115 0.058
lstm.cesm2.ssp585 15.92% 0.952 0.737 0.346 0.159 1.256 0.033 0.060 0.024
nv.cesm2.ssp370 15.97% 0.850 0.589 0.342 0.510 1.002 0.032 0.131 0.106
xgboost.mri_esm2_0.ssp434 16.28% 0.898 0.588 0.294 0.352 1.185 0.030 0.125 0.059
cnn.cesm2.ssp585 16.32% 0.923 0.693 0.322 0.165 1.270 0.036 0.087 0.042
lstm.ec_earth3.ssp434 -16.52% 0.992 0.908 0.295 0.076 0.836 0.035 0.086 0.014
nv.ec_earth3.ssp434 -17.81% 0.864 0.663 0.335 0.531 0.704 0.057 0.173 0.080
lstm.cesm2.ssp245 18.68% 0.925 0.701 0.339 0.154 1.259 0.039 0.062 0.030
lstm.cesm2.ssp370 19.37% 0.948 0.703 0.354 0.194 1.294 0.041 0.060 0.037
xgboost.cesm2.ssp585 19.59% 0.876 0.609 0.309 0.328 1.205 0.036 0.105 0.057
cnn.cesm2.ssp245 20.33% 0.888 0.646 0.308 0.228 1.286 0.042 0.090 0.052
cnn.cesm2.ssp370 20.98% 0.902 0.651 0.323 0.178 1.315 0.043 0.094 0.054
xgboost.cesm2.ssp245 22.96% 0.841 0.581 0.302 0.328 1.208 0.042 0.106 0.063
xgboost.cesm2.ssp370 24.20% 0.852 0.581 0.320 0.343 1.244 0.044 0.104 0.064

Time series of the first days

How Often Peaks Hit Hourly

QQ Plot

Distribution of the undownscaled value on days with estimated extremes values.

On the x-axis we have the daily mean (standardized). It says Undownscaled value, but is the daily mean after the downscaling. A good idea is to plot the original undownscaled value.

The purpose of this plot is to illustrate the distribution of P(undownscaled value | we predicted an extreme). This is useful because it reveals how much information we can recover concerning extreme events. If the distribution is skewed to the right, it suggests that we’re predicting extreme values only when extreme values have already occurred. Conversely, if the lower tail of the distribution resembles the reanalysis data, it indicates that we can capture short-duration extremes (e.g., brief periods of heavy rainfall, such as an intense downpour lasting an hour before stopping).

Autocorrelogram

Extremogram